Characterising Planktonic Dinoflagellate Diversity in Singapore Using DNA Metabarcoding

Metabarcoding and Metagenomics 2: 1–14 DOI 10.3897/mbmg.2.25136

Research Article

Characterising planktonic dinoflagellate diversity in Singapore using DNA metabarcoding

Yue Sze1, Lilibeth N. Miranda2, Tsai Min Sin2,†, Danwei Huang1,2

1 Department of Biological Sciences, National University of Singapore, Singapore 117558, Singapore. 2 Tropical Marine Science Institute, National University of Singapore, Singapore 119227, Singapore. † Deceased.

Corresponding author: Danwei Huang ([email protected])

Academic editor: Thorsten Stoeck | Received 19 March 2018 | Accepted 24 April 2018 | Published 17 May 2018

Abstract Dinoflagellates are traditionally identified morphologically using microscopy, which is a time-consuming and labour-intensive process. Hence, we explored DNA metabarcoding using high-throughput sequencing as a more efficient way to study planktonic dinoflagellate diversity in Singapore’s waters. From 29 minimally pre-sorted water samples collected at four locations in western Singapore, DNA was extracted, amplified and sequenced for a 313-bp fragment of the V4–V5 region in the 18S ribosomal RNA gene. Two sequencing runs generated 2,847,170 assembled paired-end reads, corresponding to 573,176 unique sequences. Sequenc- es were clustered at 97% similarity and analysed with stringent thresholds (≥150 bp, ≥20 reads, ≥95% match to dinoflagellates), recovering 28 dinoflagellate taxa. Dinoflagellate diversity captured includes parasitic and symbiotic groups which are difficult to identify morphologically. Richness is similar between the inner and outer West Johor Strait, but variations in community structure are apparent, likely driven by environmental differences. None of the taxa detected in a recent phytoplankton bloom along the West Johor Strait have been recovered in our samples, suggesting that background communities are distinct from bloom communities. The voluminous data obtained in this study contribute baseline information for Singapore’s phytoplankton communities and prompt future research and monitoring to adopt the approach established here.

Key Words high-throughput sequencing; Johor Strait; molecular operational taxonomic unit; phytoplankton; Singapore Strait

Introduction selves (Taylor et al. 2008). As part of the phytoplankton community, they are primary producers that form the base Dinoflagellates (Alveolata: Dinophyceae) are a diverse of food chains and are a large constituent of the aquatic and abundant group of unicellular protists found in both system’s food web (Spector 1984; Taylor et al. 2008). marine and freshwater environments. With more than In the marine environment, there are more than 1,500 2,000 living species described (Gómez 2012), they are one species including both photosynthetic and non-photosyn- of the most functionally important organisms in a variety thetic types (Taylor et al. 2008, Gómez 2012). Massive of aquatic ecosystems (Spector 1984; Taylor et al. 2008). proliferation and accumulation of planktonic dinoflagellates On top of being species rich, they have highly diversified can lead to water colouration and certain species are also morphologies, physiologies and biochemical properties known to cause damage to marine ecosystems as harmful (Hackett et al. 2004; Taylor et al. 2008). Dinoflagellates algal blooms or ‘red tides’ (Berdalet et al. 2016). Fish kills can be armoured or unarmoured (naked), based on the and other animal mortalities are common especially when presence or absence of thecal plates respectively (Netzel the bloom involves toxin-producing dinoflagellates such as and Dürr 1984). They play diverse ecological roles—as Karenia brevis (Landsberg et al. 2009). Incidents of paralyt- autotrophs, heterotrophs or mixotrophs—and can be endo- ic shellfish poisoning by dinoflagellates such as Alexandri- symbionts of other organisms, notably the shallow-water um fundyense have serious health implications higher up the reef-building corals, or host other endosymbionts them- food chain especially where human consumption is involved

Copyright Yue Sze et al. This is an open access article distributed under the terms of the Creative Commons Attribution License (CC BY 4.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. 2 Sze et al.: Characterising planktonic dinoflagellate diversity in Singapore using DNA metabarcoding

(Tan and Lee 1986; John et al. 2015). For example, in 2008, simultaneously, rapidly and cost-effectively (Aylagas et al. an intense bloom in the St Lawrence Estuary was found to 2016). Important ecological applications of metabarcoding be responsible for the death of many fish, bird and mammal include diet and biodiversity analyses, with environmen- species and extremely hazardous levels of paralytic shell- tal samples of various ecosystems obtained mainly from fish toxins produced by the dinoflagellate were detected in faeces (Srivathsan et al. 2015; Guillerault et al. 2017), soil edible mussels collected during the event (Starr et al. 2017). (Andersen et al. 2012; Treonis et al. 2018) or water sam- The most common and earliest method of observing ples (Thomsen et al. 2012; Yamamoto et al. 2017). phytoplankton is through light microscopy, which is time In the marine environment, metabarcoding has been consuming, labour intensive and demands a high level useful for studying both benthic and planktonic eukary- of taxonomic expertise (Sinigalliano et al. 2009; Medlin otic diversity (Fonseca et al. 2010, 2014; Logares et al. 2013; McNamee et al. 2016). Some species, such as those 2014; de Vargas et al. 2015; Guardiola et al. 2015; Le in Alexandrium Halim, 1960, are relatively featureless Bescot et al. 2016; Tragin et al. 2018). Studies have tar- and furthermore belong to species complexes, rendering geted the 18S rRNA and successfully characterised phy- them challenging to identify morphologically (John et al. toplankton communities, which are more diverse than 2005; Anderson et al. 2012; John et al. 2014). In addition, previously thought (de Vargas et al. 2015; Le Bescot et naked dinoflagellates are also difficult to identify from al. 2016). Small and symbiotic species have escaped tra- preserved samples. Hence, more efficient approaches to ditional microscopic detection but can now be identified study dinoflagellates have been developed, including the via metabarcoding (Le Bescot et al. 2016, Leblanc et al. use of scanning electron microscopy (Jung et al. 2010), 2018). Here we use this technique to study phytoplankton natural fluorescence (Karlson et al. 2010), flow cytom- communities in the coastal waters of Singapore. etry (Sinigalliano et al. 2009), high performance liquid The earliest studies of Singapore’s marine phytoplank- chromatography (HPLC) (Gin et al. 2006) and molecular ton communities were carried out in the mid-1900s, fo- methods (Karlson et al. 2010). cusing on the Singapore Strait (Tham 1953, 1973; see Molecular techniques exploit the genetic distinction Lee et al. 2015). It has long been hypothesised that the between species to identify and quantify species and have succession of phytoplankton species is likely caused by played an integral role in our understanding of the sys- the biological history of water, stability and turbulence tematic positions and evolutionary relationships amongst of the water column and monsoon-driven currents (Tham organisms (Karlson et al. 2010). Examples of earlier mo- 1973). There have also been new species discoveries lecular methods include cloning and sequencing (Zard- (Holmes 1998) and new species records for Singapore oya et al. 1995), as well as heteroduplex mobility assay (Leong et al. 2015). More recently, studies have begun (Oldach et al. 2000). More recently, DNA barcoding has to use more advanced techniques, which include the use been successful in identifying dinoflagellates (Stern et al. of extracted chlorophyll measurements (Gin et al. 2000), 2010). It uses the amplified sequence from a short, stan- flow cytometry (Gin et al. 2003), HPLC pigment analysis dardised DNA locus as diagnostic characters for species (Gin et al. 2003, 2006), spectral fluorometric characteri- identification (Lin et al. 2009; Shokralla et al. 2012). sation (Kuwahara and Leong 2015) and also molecular Loci tested for dinoflagellates include the nuclear small methods (Tang et al. 2007; Leong et al. 2015). The con- ribosomal subunit (18S rRNA) (Lin et al. 2006), large ri- sensus of these efforts is that, amongst all phytoplank- bosomal subunit (28S rRNA; D1/D2 region) (Elferink et ton groups, diatoms have the highest abundance except al. 2017), internal transcribed spacers (Stern et al. 2012) during dinoflagellate blooms (Gin et al. 2006; Leong et and the mitochondrial cytochrome c oxidase subunit I al. 2015). Particularly in the Johor Strait, there appears and cytochrome b (Lin et al. 2009). The mitochondrial to be an inverse relationship between dinoflagellate and genes cannot be easily amplified from all dinoflagellate diatom counts (Chia et al. 1988). taxa (Stern et al. 2012) and, amongst the nuclear loci, 18S There have been more than 20 positively identified spe- rRNA remains the most well-sequenced marker in public cies of dinoflagellates reported from Singapore’s waters, databases due to their widespread use in dinoflagellate but those which are known in detail have been studied systematics (Mordret et al. 2018). mainly because they are associated with harmful algal Improved technologies for performing high-through- blooms (Leong et al. 2015). These toxin-producing dino- put sequencing have enabled multiplexed amplification flagellate species are commonly associated with human products to be sequenced from numerous samples or from illnesses such as paralytic shellfish poisoning and -diar an environmental sample, in a method generally known rhetic shellfish poisoning (Holmes and Teo 2002). Based as DNA metabarcoding, which is meeting the needs of on phylogenetic analysis and identification using the large many ecologists for rapid taxon identification (Metzker subunit rDNA (28S rDNA), Tang et al. (2007) presented 2010; Taberlet et al. 2012). DNA extracts can be amplified the first evidence of icthyotoxin production by Alexandri- using primers tagged with a short barcode which labels um leei and showed that A. leei strains in Singapore were the amplicons according to the sampling design and all more similar to isolates from Malaysia than to a strain from amplicons can be pooled for sequencing and data traced Korea. More recently, Leong et al. (2015) isolated clones back to individual samples (Schnell et al. 2015). DNA me- of the 18S rDNA that were sequenced and compared to tabarcoding enables thousands of samples to be sequenced publicly available data from the GenBank nucleotide da-

https://mbmg.pensoft.net Metabarcoding and Metagenomics 2: e25136 3 tabase. With this method and morphological confirmation, 3’) and reverse primer, REL18S1R (5’–AAC AAA TCC they detected three new dinoflagellate records for Singa- AAG AAT TTC ACC TCT GAC–3’), which is the reverse pore—Karenia mikimotoi, Karlodinium cf. australe and complement of the published Dino18SF3 designed specifi- Karlodinium cf. veneficum (Leong et al. 2015). cally for dinoflagellates (Lin et al. 2006; see also Ki 2012). While more techniques are being explored, This primer combination was designed based on highly high-throughput sequencing has yet to enter the main- conserved priming sites that could amplify a variable frag- stream of Singapore’s phytoplankton research, especially ment, using a MAFFT version 7.222 (Katoh et al. 2002; for the non-bloom baseline phytoplankton communities. Katoh and Toh 2008; Katoh and Standley 2013) alignment Hence, the aim of this study is to take a metabarcoding of known dinoflagellate species from Singapore (Leong et approach to characterise the dinoflagellate communities al. 2015). The fragment length chosen was aimed at maxi- at four sites along the western coast of Singapore (Fig. mising the overlap between the two paired-end reads from 1). We seek to understand the efficacy of this approach the Illumina MiSeq System. A unique 8-bp barcode gen- to recover dinoflagellate taxa that are typically identified erated using Barcode Generator 2.8 (Comai and Howell microscopically and even detect species new to this local- 2009) was added to the 5’ end of each primer to pool sam- ity. The data also allow a preliminary analysis of spatial ples for multiplexed sequencing (Suppl. material 1). and temporal variations in community composition along Three PCR replicates were carried out for each sample the West Johor Strait (WJS), to test the hypothesis that with a 25-μl reaction mixture containing 2.5 μl 10× re- mixing between the inner and outer WJS homogenises action buffer, 2.0 μl dNTPs, 1.0 μl of 10μM forward and communities along the Strait. Overall, this DNA metabar- reverse primers, 0.2 μl Bioready rTaq (Bulldog Bio), 2 μl coding study advances our understanding of dinoflagel- DNA diluted 5× or 10× and 16.3 μl water. For each unique late diversity and distribution and provides a baseline primer combination, a negative control (without DNA) against which monitoring of phytoplankton communities was also prepared. In other words, every PCR for an ac- in Singapore can be performed. tual sample was accompanied by a negative control using the same pair of barcoded primers. During the analyses, MOTUs appearing in the negative controls post-filtering Material and Methods would be removed from their corresponding samples. The PCR protocol comprised 1 min of initial denaturation at Study sites and sampling 94 °C, followed by 35 cycles of 45 s at 94 °C, 45 s at 53 A total of 29 plankton samples were collected between Sep- °C and 1 min at 72 °C, ending with 3 min at 72 °C. Am- tember and December 2015 at four sites in western Singapore plified products were quantified using the Qubit dsDNA (Fig. 1). Along the West Johor Strait (WJS), three replicate BR Assay Kit on a Qubit 3 Fluorometer (Thermo Fisher samples per month were collected from stations at inner and Scientific) in order to pool approximately equal amount outer WJS (1.45883°, 103.7202° and 1.34037°, 103.63018°, of PCR product for each tagged amplicon. Purification of respectively; 24 samples). Along the Singapore Strait, two the mixed products was performed using SureClean Plus samples were collected from St John’s Island (SJI) and three (Bioline) following the manufacturer’s protocol. at a station off Jurong Island (JI) (1.22247°, 103.8487° and The pooled PCR products, including all negative con- 1.29527°, 103.68202°, respectively; 5 samples). trols, were split into two for Illumina DNA library prepa- Collections were carried out using a 15-μm-mesh ration and paired-end sequencing with the Illumina MiSeq plankton net hauled vertically from 5 m depth to the wa- System. Approximately half of a sequencing run was target- ter surface. From each haul, the volume of water collected ed for each library—the first run generated read lengths of was standardised using a 50-ml measuring cylinder and 2× 250 bp (MiSeq Reagent Kit v2), while the second 2× temporarily transferred to a plastic bottle. Each sample 300 bp (MiSeq Reagent Kit v3). was filtered through an 8-µm cellulose filter paper (What- man, Sigma-Aldrich), which was immediately wrapped Bioinformatic pipeline in sterilised aluminium foil, snap frozen in liquid nitrogen and stored at -30 °C prior to DNA extraction. Assembly of paired-end reads was performed using Paired-End reAd mergeR (PEAR) version 0.9.6 (Zhang DNA extraction, amplification and sequencing et al. 2014), with the criteria of 100-bp minimum overlap, minimum and maximum lengths at 150 and 330 bp re- Each filter paper was cut into two and placed in separate spectively and Phred quality threshold of 30. Amongst all 1.5-ml tubes for DNA extraction using the standard CTAB the dinoflagellate sequences from GenBank that were an- (cetyltrimethylammonium bromide) and phenol-chloro- alysed here, few indels were found internal to the priming form protocol (Doyle and Doyle 1987). DNA pellets were regions, with minimum and maximum lengths of 248 and re-suspended in 100 μl of water, pooled between the two 263 bp respectively. Therefore, our filtering criterion with tubes and stored at -30°C until DNA amplification. respect to read length would capture all dinoflagellates. A 313-bp, V4-V5 region of the 18S rRNA locus was am- The assembled reads were analysed in OBITools plified using newly-designed forward primer, REL18S1F version 1.2.0 (Boyer et al. 2016). A maximum of 2-bp (5’– GTT GCG GTT AAA AAG CTC GTA GTT GGA– mismatch for primer tags was used to assign sequence

https://mbmg.pensoft.net 4 Sze et al.: Characterising planktonic dinoflagellate diversity in Singapore using DNA metabarcoding

Figure 1. Sampling stations where plankton was collected for this study between September and December 2015. Arrow represents predominant current direction, stronger during the northeast monsoon (November to March) and weaker during the southwest monsoon (June-September). records to corresponding samples with ngsfilter. obian- Data analyses notate followed by obisplit re-annotated sequence de- scriptions and separated them into files based on their as- We compared dinoflagellate MOTU identities across the signed samples. Within each sample file, strictly identical PCR triplicates and only retained those that met the above sequences were grouped together and assigned a count criteria in at least two replicates. MOTUs appearing in number with obiuniq. obiclean was used to detect am- the negative controls were to be removed from their cor- plification or sequencing errors, by classifying sequence responding sample set if they met the same criteria. Re- records to either ‘head’, ‘internal’ or ‘singleton’, taking tained sequences were combined first according to their into account sequence similarities and record counts. sample and then site, noting their respective number of The ‘head’ is the most common sequence amongst all reads for each MOTU. Sequences from all samples were sequences, while a ‘singleton’ sequence is one with no combined to determine the total number of unique dino- other variants in the amplification product (Guardiola flagellate MOTUs amongst all samples. et al. 2015); ‘internal’ is neither of the above and most Phylogenetic analyses were carried out to determine likely corresponds to a sequencing or amplification error. relationships amongst the MOTUs and previously se- Only ‘head’ sequences of at least 150 bp were retained quenced taxa. Rhodophytes Rhodella violacea (Korn- for further analysis using obigrep and obisplit. mann) Wehrmeyer (GenBank accession AF168624) Sequences with at least 20 reads were clustered into and Bangia atropurpurea (Mertens ex Roth) C.Agardh molecular operational taxonomic units (MOTUs) with (AF169339) were selected as outgroups. Taxa analysed in USEARCH version 8.1.1861 (Edgar 2010) at a similarity Leong et al. (2015) as well as published sequences from threshold of 97% and the greedy clustering algorithm. For GenBank that matched ≥95% to MOTUs obtained here each cluster, the sequence with the highest number of reads were also included to construct a data matrix with 100 was taken as the ‘centroid’, where counts of the remaining terminals. Sequences were aligned using MAFFT version sequences with at most 3% dissimilarity were added to it. 7.222 (Katoh et al. 2002; Katoh and Toh 2008; Katoh Chimeric sequences were also removed in this process. For and Standley 2013). A neighbour-joining (NJ) tree was taxonomic assignment, the representative sequence for each inferred using PAUP* version 4.0b10 (Swofford 2003), MOTU was searched against the GenBank sequence data- with branch supports assessed using 1,000 bootstrap base using BLAST+ version 2.2.31 (Altschul et al. 1990). pseudoreplicates (Felsenstein 1985). We also performed Only sequence matches with minimum identity of 95% to model-based maximum likelihood (ML) and Bayesian any dinoflagellate were retained for further analyses. analyses. ML tree searches via RAxML version 8.0.9 https://mbmg.pensoft.net Metabarcoding and Metagenomics 2: e25136 5

(Stamatakis 2006, 2014; Stamatakis et al. 2008) were car- species and were not nested within a known clade include ried out with 50 replicates, default GTRGAMMA substi- OTU12, OTU45, OTU58, OTU96, OTU97 and OTU102. tution model and 1,000 bootstrap replicates. For Bayesian High genetic diversity was observed amongst the MO- analysis, we selected the most suitable substitution mod- TUs nested within Gyrodinium spirale and G. fusiforme el using jModelTest 2.1.10 (Guindon and Gascuel 2003; (OTU1, OTU64, OTU110, OTU123 and OTU127). Posada 2008; Darriba et al. 2012) based on the Akaike in- Some MOTUs could be placed in a likely taxon, such as formation criterion (AIC). MrBayes 3.2.6 (Huelsenbeck OTU20 in Gymnodinium and OTU129 in Prorocentrum, and Ronquist 2001; Ronquist and Huelsenbeck 2003; as they had high sequence similarity with and were nested Ronquist et al. 2012) was used to generate four runs of within known taxa. There were also MOTUs exhibiting 11 million Markov chain Monte Carlo iterations each, exact sequence matches to specific dinoflagellate species, saving a tree every hundredth iteration. Convergence an indication that such species were already known. These amongst runs was determined using Tracer 1.6 (Rambaut were represented by zero branch length difference -be et al. 2014), which led to the removal of the first 50,001 tween the MOTU and the known species, such as OTU23 trees as burn-in. with Amphidinium klebsii, OTU73 with Alexandrium cohorticula and OTU81 with Gyrodinium instriatum. None of the samples recovered bloom-forming dinoflagel- Results lates such as Karlodinium and Takayama, which were detected in abundance during a recent phytoplankton bloom along The two Illumina MiSeq sequencing runs produced the WJS in February 2014 (Lim et al. 2014; Leong et al. 2,847,170 assembled paired-end reads corresponding 2015). OTU98 was closely related to Symbiodinium sequenc- to 573,176 unique sequences. Further error pruning, se- es and was also observed during the bloom but could be part quence filtering and chimera removal further reduced the of the background community as ex hospite zooxanthellae. number of unique sequences to 268. After clustering the The two most widespread MOTUs were OTU1 (iden- sequences using a 97% similarity threshold, 133 MOTUs tical sequence with G. fusiforme) and OTU20 (uncultured were recovered. Filtering based on ≥95% sequence simi- dinoflagellate). The former was present at both WJS sites larity to GenBank dinoflagellate sequences resulted in 28 and Jurong Island, while the latter was detected at both unique MOTUs remaining (Table 1). Sequences have been WJS sites and St John’s Island (Table 1). None of the deposited into GenBank (accession numbers MH234223– MOTUs were found at all sites. OTU1 was detected at MH234250). The mean number of reads per sample was WJS during every sampling and registered consistently 46,386 (± S.E. 9,743) and the mean number of MOTUs per high read count across sites—as high as 10,000× the next sample was 3.79 (± S.E. 0.45). At the site level, considering most abundant taxon (i.e. September 2015 at inner WJS). only stations with three samples collected, there were on Nearly half of the MOTUs (13 of 28) were detected at average 4.56 (± SE 1.25) dinoflagellate MOTUs per sam- inner and outer WJS each. There was an overlap of four pling station (Table 1). Our negative controls contained an MOTUs, including the abundant OTU1 (G. fusiforme), average of 93 sequences per sample, but none of these met between the two WJS sites. Fewer MOTUs were found the set criteria (≥150 bp, ≥20 reads, ≥95% match). at Jurong Island (6 of 28) and St John’s Island (8 of 28). The phylogenetic analyses revealed deep relationships Two-thirds of the former’s MOTUs were unique to Ju- that were generally inconsistent amongst NJ, ML and rong Island, while most of St John’s Island’s MOTUs (6 Bayesian reconstructions, but these were poorly support- of 8) were found at WJS. Jurong Island and St John’s Is- ed across all analyses. There were no topological con- land were sampled only at one time point, so results ought flicts that were supported by any of the analyses, so we to be viewed with caution. At WJS, the highest MOTU discuss the well-supported nodes using the NJ tree (Fig. richness was observed during October 2015 (Fig. 3). Par- 2). Branch supports increased closer to the tips, thus help- ticularly for outer WJS, 12 of the 13 MOTUs were detect- ing to corroborate most of the BLAST matches. Of the 28 ed during that month, but apart from October, the site had MOTUs, 14 were matched 100% to a known sequence very low MOTU richness. from GenBank and most of these were strongly supported on the tree (NJ, ML bootstrap ≥90, and Bayesian posteri- or probability =1; e.g. OTU61 + uncultured dinoflagellate Discussion GU819712, and OTU31 + Gonyaulax spinifera). OTU10, OTU37 and OTU61 showed no sequence The two Illumina MiSeq sequencing runs generated a similarity to any known species but exhibited high simi- large number of reads that likely represent the majority larity to a limited set of ‘uncultured dinoflagellates’ (Ta- of dinoflagellate individuals captured on the sample fil- ble 1). They were phylogenetically distinct from the rest ters. Overall, the use of DNA metabarcoding on minimal- of the dinoflagellates (Fig. 2) and considered to- beas ly sorted water samples has recovered a greater diversity sociated with Syndiniales, belonging to the deep-branch- of dinoflagellates in Singapore than a previous study that ing marine alveolates (MALV; Lopez-García et al. 2001; sequenced clone isolates amplified from water samples Moon-van der Staay et al. 2001; Guillou et al. 2008). Oth- at comparable sites (Leong et al. 2015). As the diversity er MOTUs that were not genetically similar to any known estimates here are based principally on sequence analy-

https://mbmg.pensoft.net 6 Sze et al.: Characterising planktonic dinoflagellate diversity in Singapore using DNA metabarcoding

Figure 2. Neighbour-joining (NJ) tree showing 18S rRNA sequence relationships amongst dinoflagellates in Singapore (bold) and from GenBank. The 28 molecular operational taxonomic units detected in this study are denoted by prefix ‘OTU’. Dinoflagellate orders are represented by coloured branches. Bootstrap supports based on NJ and maximum likelihood (ML) methods (≥50), as well as Bayesian posterior probabilities (≥0.8) are shown as circles at the nodes. sis, they need to be validated with other high-throughput sequences are in GenBank varies amongst genera. Se- methods such as flow cytometry and morphological ex- quences from the genus Alexandrium are most abundant amination using, for instance, scanning electron micros- on GenBank, with approximately 500 available. This is copy. Nevertheless, the spatial and temporal variations followed by Prorocentrum with 97 sequences retrieved; of community composition, shown here, form essential all other genera are represented by fewer sequences, with hypotheses for further tests of local and regional dinofla- some having only one sequence available. Despite having gellate distribution patterns. the largest collection of published 18S rRNA sequences After retaining sequences with ≥95% similarity to available, sequencing effort for dinoflagellates is highly GenBank dinoflagellate sequences, only 28 dinoflagellate skewed amongst genera. The majority of MOTUs, not as- MOTUs have been detected. It is worth noting that the signed to dinoflagellates, indicate that the primers which extent of how well represented 18S rRNA dinoflagellate have been designed for dinoflagellates also capture oth-

https://mbmg.pensoft.net Metabarcoding and Metagenomics 2: e25136 7

Table 1. Molecular operational taxonomic units of dinoflagellates recovered from four sites along the western coast of Singapore, with information about their closest GenBank matches and sequencing read counts at each site and sampling month in 2015.

GenBank Outer West Johor Inner West Johor Strait Jurong St John’s MOTU accession GenBank closest match Identity Strait Island Island number Sep Oct Nov Dec Sep Oct Nov Dec AB120002 Gyrodinium fusiforme OTU1 MH234223 100.0 367351 167408 357281 226843 147 4302 688 185 96020 (Takano and Horiguchi 2004) GU819712 Uncultured dinoflagellate OTU10 MH234224 100.0 1036 476 6114 320 1268 (Edgcomb et al. 2011) AB827556 Uncultured dinoflagellate OTU11 MH234225 97.7 526 1315 (Kok et al. 2014) KC488405 Uncultured dinoflagellate OTU12 MH234226 98.5 141 4347 99 (Dasilva et al. 2014) AJ968729 Paulsenella vonstoschii OTU16 MH234227 100.0 4435 (Kühn and Medlin 2005) OTU20 MH234228 FJ914425 Uncultured dinoflagellate 100.0 395 4362 139 502 OTU23 MH234229 EU046335 Amphidinium klebsii 100.0 691 1968 OTU31 MH234230 FR865625 Gonyaulax spinifera 100.0 1415 GU819660 Uncultured dinoflagellate OTU37 MH234231 100.0 813 (Edgcomb et al. 2011) AB181899 Protoperidinium pallidum OTU38 MH234232 95.4 561 (Yamaguchi and Horiguchi 2005) EU780628 Uncultured eukaryote OTU45 MH234233 99.6 257 113 (Guillou et al. 2008) OTU48 MH234234 FJ914426 Uncultured dinoflagellate 98.5 619 OTU51 MH234235 FJ467492 Warnowia sp. (Gómez et al. 2009) 100.0 32 AB694525 Uncultured dinoflagellate OTU58 MH234236 99.2 147 77 (Kok et al. 2012) GU820302 Uncultured dinoflagellate OTU61 MH234237 100.0 162 23 (Edgcomb et al. 2011) AB120002 Gyrodinium fusiforme OTU64 MH234238 95.9 503 (Takano and Horiguchi 2004) OTU72 MH234239 AJ276699 Ceratium furca 99.6 145 48 AF113935 Alexandrium cohorticula OTU73 MH234240 100.0 130 (Usup et al. 2002) OTU81 MH234241 AY721981 Gyrodinium instriatum 100.0 36 FN598427 Uncultured dinoflagellate OTU96 MH234242 100.0 21 (Sauvadet et al. 2010) GU819784 Uncultured dinoflagellate OTU97 MH234243 99.6 21 21 (Edgcomb et al. 2011) FJ823465 Uncultured Symbiodinium OTU98 MH234244 100.0 41 (Annenkova et al. 2011) AB827518 Uncultured dinoflagellate OTU102 MH234245 97.7 22 (Kok et al. 2014) OTU107 MH234246 FJ467492 Warnowia sp. (Gómez et al. 2009) 98.1 466 AB120002 Gyrodinium fusiforme OTU110 MH234247 98.8 21 (Takano and Horiguchi 2004) AB120002 Gyrodinium fusiforme OTU123 MH234248 96.5 380 (Takano and Horiguchi 2004) AB120002 Gyrodinium fusiforme OTU127 MH234249 98.0 34 62 (Takano and Horiguchi 2004) EF492510 Prorocentrum mexicanum (Leblond et al. 2010) EF492511 Prorocentrum micans (Leblond et al. 2010) OTU129 MH234250 JQ616822 Prorocentrum rhathymum 100.0 147 (Herrera-Sepúlveda et al. 2013) AJ415520 Prorocentrum minimum JQ390504 Prorocentrum texanum (Henrichs et al. 2013)

https://mbmg.pensoft.net 8 Sze et al.: Characterising planktonic dinoflagellate diversity in Singapore using DNA metabarcoding er organisms. Our database searches have indeed recov- taxon assignments based on the 18S rRNA gene may not ered several diatoms amongst other eukaryotes. The 18S be precise, this species is known to be widespread and has region amplified falls within the V4–V5 region that is been recorded in Indonesia and the Malacca Strait (Dodge commonly used to broadly amplify eukaryotic 18S rDNA 1982; Noor et al. 2007). Present at the inner and outer WJS using universal primers (Hadziavdic et al. 2014). Despite every month of sampling from September to December using primer sequences that appear to be specific to di- 2015, it is one of only four shared MOTUs at the two sites. noflagellates, the PCRs have also allowed non-specific Interesting, G. fusiforme was not detected during the 2014 binding. Future studies could consider adjacent priming Karlodinium australe bloom on the Singapore side of the sites to reduce non-target amplification. channel (Leong et al. 2015). Furthermore, Gyrodinium As with the previous study in Singapore by Leong et was only one of several dinoflagellate taxa constituting a al. (2015), our high-throughput sequencing recovers par- minute proportion (<0.2%) of cell densities observed on asitic and symbiotic dinoflagellates which are challeng- the Malaysian side (Lim et al. 2014). With the exception ing to detect using conventional microscopic sorting. The of the MALV and Symbiodinium, there appears to be mu- three MOTUs nested in the Syndiniales or MALV cluster tual exclusivity between bloom and background dinofla- (OTU10, OTU37 and OTU61) are clear outgroups of the gellate communities (Fig. 2). In particular, Takayama and ‘core’ dinoflagellate clade (Guillou et al. 2008; Taylor Karlodinium, genera linked to mass fish mortality during et al. 2008). While this and previous studies (e.g. Guil- the bloom (Leong et al. 2015), were not detected from our lou et al. 2008; Leong et al. 2015) find Syndiniales to be samples. These is possibility that they have been excluded monophyletic, as a whole, it may be paraphyletic (Stras- from the inner WJS due to limited exchange with the Sin- sert et al. 2018). Syndiniales are obligate parasites found gapore Strait which connects to other regions. However, throughout marine habitats from the water surface to sed- as they were also absent from the outer WJS, Jurong Is- iments (Guillou et al. 2008) and are often highly repre- land and St John’s Island, it is more likely that they were sented in environmental samples (Kok et al. 2012). They encysted and benthic during our sampling, although cysts have extremely diverse host ranges, associating with not have yet to be observed for these species (Bergholtz et al. only alveolates, but also copepods, cnidarians and even 2005; Wang et al. 2011). Sequencing of benthic sediments fish eggs (Chambouvet et al. 2008). Even if they are in from the WJS may help detect and identify cysts that can low abundance, the community could quickly increase be matched to the vegetative forms in the water column in population size and affect the abundance of its host, (Godhe et al. 2002; Nagai et al. 2012; Gao et al. 2017). consequently impacting ecosystem functioning (Logares Contrary to the hypothesis that mixing between the in- et al. 2015). It has only been recorded recently in Singa- ner and outer WJS homogenises the communities along pore’s waters during the 2014 bloom (Leong et al. 2015). the Strait, considerable differences have been found be- Our results show that read counts of sequences closely tween the two sampling stations (Fig. 3; Table 1). The related to Syndiniales are relatively high (Table 1)—an West Johor Strait is a shallow channel bounded by the indication that these marine parasites are prevalent in our Causeway and Pulai River (Kazemi et al. 2014). Species waters even in non-bloom conditions. compositional differences between the inner and outer A symbiotic dinoflagellate of the genus Symbiodinium WJS suggest that there are site-specific variations related (OTU98) has been detected at inner WJS in December to coastal hydrodynamics and physico-chemical proper- 2015. That this taxon is not more widespread and even ties. Surrounding water bodies and monsoonal patterns absent in samples from the reef environment of St John’s determine much of the local hydrodynamics and water Island is surprising because of its ubiquity and importance parameters such as salinity, temperature and turbidi- as endosymbionts of scleractinian corals (zooxanthellae; ty (Behera et al. 2013). The circulation of Singapore’s Tanzil et al. 2016), other marine invertebrates and protists waters is largely driven by two dominant monsoon sea- (Takabayashi et al. 2012). They can exist as free-living sons—the northeast (November to March) and southwest cells that are detectable outside the hosts, sometimes in (June–September) monsoons separated by two brief in- high densities and serve as symbiont sources to be taken ter-monsoon periods (Behera et al. 2013). Our sampling up by juvenile corals or for uptake by adult corals faced period coincides only briefly with the two monsoons and with environmental stress (Littman et al. 2008; Takaba- the four-month sampling period is insufficient for mak- yashi et al. 2012). While it is not impossible to identify ing meaningful temporal comparisons. Nevertheless, the them using microscopy alone, they are morphologically temporal variations appear to be related to flushing po- similar to other dinoflagellates and have often been mis- tential. Tidal mixing would reduce stratification and the classified (Littman et al. 2008). Therefore, DNA-based hydrodynamic response to variation in freshwater inflows methods are useful for distinguishing Symbiodinium from is dependent on the flushing time of the channel (Kazemi other dinoflagellates. Knowing that they can be captured et al. 2014). The inner WJS is known to have low flushing in our study, future research or monitoring programmes potential with a high residence time of >70 days (Bay- can utilise DNA metabarcoding to help track the pool of en et al. 2013) and is also affected by consistently high potential coral endosymbionts in Singapore waters. levels of run-off from the Johor rivers resulting in more By far, the most read-abundant MOTU is OTU1, which eutrophic conditions (Sin et al. 2016). These factors could has a sequence identical to Gyrodinium fusiforme. While drive the slightly more uniform community structure and

https://mbmg.pensoft.net Metabarcoding and Metagenomics 2: e25136 9

Figure 3. Number of dinoflagellate molecular operational taxonomic units (MOTUs) recovered at each of four sites, inner West Johor Strait (Inner WJS, red), outer West Johor Strait (Outer WJS, orange), Jurong Island (JI, green) and St John’s Island (SJI, blue), as well as the month of sampling in 2015. diversity over time at the inner WJS compared to the out- across all living cells. It comprises nine hypervariable re- er WJS which, being closer to the opening of the channel, gions (V1–V9) (Ki 2012; Hadziavdic et al. 2014), with V2, would have greater tidal mixing and flushing potential. V4 and V9 being at sufficient interspecific variabilities for The primary goal of all high-throughput sequencing of biodiversity assessments (Hadziavdic et al. 2014). Never- amplified markers is to recover accurate MOTU sequenc- theless, all eight regions of eukaryotes (lacking V6) have es and richness estimates from the voluminous sequence been targeted for sequencing (Ki 2012). Here, in order to data (Nguyen et al. 2015). The number of MOTUs is de- place our MOTUs in the same phylogenetic context with pendent on the similarity threshold applied to sequence known dinoflagellates from Singapore, we designed prim- reads for delimiting taxonomic units. Sequences that are ers to target the hypervariable V4–V5 region, which is also more similar than a given threshold will be grouped into internal to the ~600-bp marker used by Leong et al. (2015) the same MOTU (Blaxter 2004). The ideal similarity val- and has a length of 313 bp for paired-end sequencing by ue used for clustering should be one that most closely ap- the Illumina MiSeq System (Hadziavdic et al. 2014; Le proximates the diversity of a sample at the species level Bescot et al. 2016; Searle et al. 2016). It is important to (Guardiola et al. 2015). At a lower similarity threshold, note that different 18S regions can lead to varying diversity fewer MOTUs could be recovered, but these then become estimates (Edgcomb and Stoeck 2012), so our results need divided into separate MOTUs when a higher similarity to be verified in future by analysing other variable regions. level is used (Logares et al. 2015). The differences in Downstream of the amplification and sequencing steps, the absolute number of MOTUs could be as much as six we have attempted to account for possible errors associat- times with clustering thresholds set at 3% and 1% dis- ed with high-throughput sequencing and DNA metabar- similarities (Logares et al. 2015) and these can vary for coding. These measures include the omission of sequenc- different genetic loci (Ki 2012). Here we have attempted es <150 bp represented by fewer than 20 reads or those clustering at various dissimilarity thresholds (1–10%) to not matching at least 95% to published dinoflagellate se- determine the sequence variabilities that show no change quences. Consequently, despite high read recovery, there in the number of clusters obtained. The suitable cut-off was low read usability, with only 268 unique sequences distance ranged from 2% to 5% amongst our samples, so that were eventually used for analysis. We also retained we used 3% as a conservative threshold so as not to over- unique sequences only if they were detected in two of estimate diversity at each site. three PCR replicates and would potentially remove signals The 18S rRNA is a commonly-used and well-charac- appearing in the negative controls. While high-throughput terised genetic marker with a highly conserved function sequencing studies such as this are efficient in producing

https://mbmg.pensoft.net 10 Sze et al.: Characterising planktonic dinoflagellate diversity in Singapore using DNA metabarcoding large amounts of sequence data, it is important to note that ‘dirt’ DNA from soil reflects vertebrate biodiversity. Molecular Ecolo- there are known issues concerning Illumina-based DNA gy 21: 1996–1979. https://doi.org/10.1111/j.1365-294X.2011.05261.x metabarcoding, including primer tag jumps, contamina- Annenkova NV, Lavrov DV, Belikov SI (2011) Dinoflagellates associ- tion and false positive detections (Esling et al. 2015). We ated with freshwater sponges from the ancient Lake Baikal. Protist have sought to minimise these effects through the detailed 162: 222–236. https://doi.org/10.1016/j.protis.2010.07.002 experimental design and by implementing the rigorous Aylagas E, Borja A, Irigoien X, Rodríguez-Ezpeleta N (2016) Bench- criteria in our bioinformatic pipeline. marking DNA metabarcoding for biodiversity-based monitoring and assessment. Frontiers in Marine Science 3: 96. https://doi. org/10.3389/fmars.2016.00096 Conclusion Bayen S, Zhang H, Desai MM, Ooi SK, Kelly BC (2013) Occurrence and distribution of pharmaceutically active and endocrine disrupting Despite the stringent thresholds, we have recovered a compounds in Singapore’s marine environment: influence of hydro- higher diversity of planktonic dinoflagellates—28 MO- dynamics and physical-chemical properties. Environmental Pollu- TUs—than previously reported for Singapore waters tion 182: 1–8. https://doi.org/10.1016/j.envpol.2013.06.028 (Leong et al. 2015). This was achieved within a short Behera MR, Chun C, Palani S, Tkalich P (2013) Temporal variability and four-month sampling period with minimal pre-sorting climatology of hydrodynamic, water property and water quality pa- and without isolation of specific organisms. The 15-μm rameters in the West Johor Strait of Singapore. Marine Pollution Bul- plankton net used for capturing microplankton would letin 77: 380–395. https://doi.org/10.1016/j.marpolbul.2013.09.043 have omitted even smaller dinoflagellates since studies Berdalet E, Fleming LE, Gowen R, Davidson K, Hess P, Backer LC, have suggested that many open water environments are Moore SK, Hoagland P, Enevoldsen H (2016) Marine harmful al- dominated by small cells in the pico- (0.8–5μm) and nano- gal blooms, human health and wellbeing: challenges and opportu- plankton (5–20 μm) ranges (Gin et al. 2000; de Vargas nities in the 21st century. Journal of the Marine Biological Asso- et al. 2015). Therefore, our estimates are conservatively ciation of the United Kingdom 96: 61–91. https://doi.org/10.1017/ low. Further studies targeting greater ranges of organism S0025315415001733 size, depth and habitat are likely to detect more taxa, both Bergholtz T, Daugbjerg N, Moestrup Ø, Fernandez-Tejedor M (2005) known and unknown. The large and yet expanding DNA On the identity of Karlodinium veneficum and description of Karlo- sequence database—to which the MOTUs obtained here dinium armiger sp. nov. (Dinophyceae), based on light and electron contribute—will enable more precise matches for DNA microscopy, nuclear-encoded LSU rDNA, and pigment composi- metabarcodes locally and in the region. tion. Journal of Phycology, 42: 170–193. https://doi.org/10.1111/ j.1529-8817.2006.00172.x Blaxter ML (2004) The promise of a DNA taxonomy. Philosophical Acknowledgements Transactions of the Royal Society of London B-Biological Sciences 359: 669–679. https://doi.org/10.1098/rstb.2003.1447 Boyer F, Mercier C, Bonin A, Le Bras Y, Taberlet P, Coissac E (2016) We thank staff and crew of the Ecological Monitoring, In- OBITools: a unix-inspired software package for DNA metabar- formatics and Dynamics Research Group (Tropical Ma- coding. Molecular Ecology Resources 16: 176–182. https://doi. rine Science Institute) and R/V Galaxea (St John’s Island org/10.1111/1755-0998.12428 National Marine Lab), for assisting with field research; Chambouvet A, Morin P, Marie D, Guillou L (2008) Control of toxic members of the Reef Ecology Laboratory for support and marine dinoflagellate blooms by serial parasitic killers. Science 322: assistance, especially Ywee Chieh Tay for advice on data 1254–1257. https://doi.org/10.1126/science.1164387 analyses; and Céline Mercier, for providing assistance re- Chia LS, Khan H, Chou LM (1988) The coastal environment profile of garding usage of OBITools. This study was funded by the Singapore. ICLARM, Manila, 1–91. National Research Foundation, Prime Minister’s Office, Comai L, Howell T (2012) Barcode Generator. Version 2.8. Available Singapore under its Marine Science R&D Programme from: http://comailab.genomecenter.ucdavis.edu/index.php/Bar- (MSRDP-P03). code_generator Cuvelier ML, Allen AE, Monier A, McCrow JP, Messie M, Tringe SG, Woyke T, Welsh RM, Ishoey T, Lee J-H, Binder BJ, DuPont CL, References Latasa M, Guigand C, Buck KR, Hilton J, Thiagarajan M, Caler E, Read B, Lasken RS, Chavez FP, Worden AZ (2010) Targeted metag- Altschul SF, Gish W, Miller W, Myers EW, Lipman DJ (1990). Basic enomics and ecology of globally important uncultured eukaryotic local alignment search tool. Journal of Molecular Biology 215: 403- phytoplankton. Proceedings of the National Academy of Sciences 410. https://doi.org/10.1016/S0022-2836(05)80360-2 USA 107: 14679–14684. https://doi.org/10.1073/pnas.1001665107 Anderson DM, Alpermann TJ, Cembella AD, Collos Y, Masseret E, Darriba D, Taboada GL, Doallo R, Posada D (2012) jModelTest 2: more Montresor M (2012) The globally distributed genus Alexandri- models, new heuristics and parallel computing. Nature Methods 9: um: multifaceted roles in marine ecosystems and impacts on hu- 772. https://doi.org/10.1038/nmeth.2109 man health. Harmful Algae 14: 10–35. https://doi.org/10.1016/j. Dasilva CR, Li WKW, Lovejoy C (2014) Phylogenetic diversity of hal.2011.10.012 eukaryotic marine microbial plankton on the Scotian Shelf North- Andersen K, Bird KL, Rasmussen M, Haile J, Breuning-Madsen H, Kjaer western Atlantic Ocean. Journal of Plankton Research 36: 344–363. KH, Orlando L, Gilbert MT, Willerslev E (2012) Meta-barcoding of https://doi.org/10.1093/plankt/fbt123 https://mbmg.pensoft.net Metabarcoding and Metagenomics 2: e25136 11 de Vargas C, Audic S, Henry N, Decelle J, Mahe F, Logares R, Lara E, Gin KYH, Lin X, Zhang S (2000) Dynamics and size structure of phy- Berney C, Le Bescot N, Probert I, Carmichael M, Poulain J, Romac toplankton in the coastal waters of Singapore. Journal of Plankton S, Colin S, Aury J-M, Bittner L, Chaffron S, Dunthorn M, Engel- Research 22: 1465–1484. https://doi.org/10.1093/plankt/22.8.1465 en S, Flegontova O, Guidi L, Horak A, Jaillon O, Lima-Mendez G, Gin KYH, Zhang S, Lee YK (2003) Phytoplankton community struc- Luke J, Malviya S, Morard R, Mulot M, Scalco E, Siano R, Vincent ture in Singapore’s coastal waters using HPLC pigment analysis F, Zingone A, Dimier C, Picheral M, Searson S, Kandels-Lewis S, and flow cytometry. Journal of Plankton Research 25: 1507–1519. Tara Oceans Coordinators, Acinas SG, Bork P, Bowler C, Gorsky G, https://doi.org/10.1093/plankt/fbg112 Grimsley N, Hingamp P, Iudicone D, Not F, Ogata H, Pesant S, Raes Gin KYH, Holmes MJ, Zhang S, Lin X (2006) Phytoplankton structure J, Sieracki ME, Speich S, Stemmann L, Sunagawa S, Weissenbach in the tropical port waters of Singapore. In: Wolanski E (Ed.) The J, Wincker P, Karsenti E, Boss E, Follows M, Karp-Boss L, Krzic U, environment of Asia Pacific harbors. Springer, The Netherlands, Reynaud EG, Sardet C, Sullivan MB, Velayoudon D (2015) Eukary- 347–375. https://doi.org/10.1007/1-4020-3655-8_21 otic plankton diversity in the sunlit ocean. Science 348: 1261605. Godhe A, Rehnstam-Holm AS, Karunasagar I, Karunasagar I (2002) https://doi.org/10.1126/science.1261605 PCR detection of dinoflagellate cysts in field sediment samples from Dodge JD (1982) Marine dinoflagellates of the British Isles. Her Majes- tropic and temperate environments. Harmful Algae 1: 361–373. ty’s Stationery Office, London, 1–303. https://doi.org/10.1016/S1568-9883(02)00053-7 Doyle JJ, Doyle JL (1987) A rapid DNA isolation procedure for small Gómez F (2012) A checklist and classification of living dinoflagellates quantities of fresh leaf tissue. Phytochemistry Bulletin 19: 11–15. (Dinoflagellata, Alveolata). CICIMAR Oceánides 27: 65–140. Edgar RC (2010) Search and clustering orders of magnitude faster than Gómez F, López-García P, Moreira D (2009) Molecular phylogeny of BLAST. Bioinformatics 26: 2460–2461. https://doi.org/10.1093/ the ocelloid‐bearing dinoflagellatesErythropsidinium and Warnowia bioinformatics/btq461 (Warnowiaceae, Dinophyceae). Journal of Eukaryotic Microbiology Edgcomb V, Orsi W, Bunge J, Jeon S, Christen R, Leslin C, Holder M, 56: 440–445. https://doi.org/10.1111/j.1550-7408.2009.00420.x Taylor GT, Suarez P, Varela R, Epstein S (2011) Protistan microbial Guardiola M, Uriz MJ, Taberlet P, Coissac E, Wangensteen OS, Turon observatory in the Cariaco Basin, Caribbean. I. Pyrosequencing vs X (2015) Deep-sea, deep-sequencing: metabarcoding extracellular Sanger insights into species richness. ISME Journal 5: 1344–1356. DNA from sediments of marine canyons. PLoS ONE 10: e0139633. https://doi.org/10.1038/ismej.2011.6 https://doi.org/10.1371/journal.pone.0139633 Edgcomb V, Stoeck T (2012) Massively parallel tag sequencing unveils Guindon S, Gascuel O (2003) A simple, fast, and accurate algorithm to the complexity of marine protistan communities in oxygen-depleted estimate large phylogenies by maximum likelihood. Systematic Bi- habitats. In: Harbers M, Kahl G (Eds) Tag-based Next Generation ology 52: 696–704. https://doi.org/10.1080/10635150390235520 Sequencing. Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim, Guillou L, Viprey M, Chambouvet A, Welsh RM, Kirkham AR, Mas- 171–183. https://doi.org/10.1002/9783527644582.ch11 sana R, Scanlan DJ, Worden AZ (2008) Widespread occurrence and Elferink S, Neuhaus S, Wohlrab S, Toebe K, Voß D, Gottschling M, genetic diversity of marine parasitoids belonging to Syndiniales (Al- Lundholm N, Krock B, Koch BP, Zielinski O, Cembella A, John U veolata). Environmental Microbiology 10: 3349–3365. https://doi. (2017) Molecular diversity patterns among various phytoplankton org/10.1111/j.1462-2920.2008.01731.x size-fractions in West Greenland in late summer. Deep Sea Research Hackett JD, Anderson DM, Erdner DL, Bhattacharya D (2004) Dinofla- Part I: Oceanographic Research Papers 121: 54–69. https://doi. gellates: a remarkable evolutionary experiment. American Journal org/10.1016/j.dsr.2016.11.002 of Botany 91: 1523–1534. https://doi.org/10.3732/ajb.91.10.1523 Esling P, Lejzerowicz F, Pawlowski J (2015) Accurate multiplexing and Hadziavdic K, Lekang K, Lanzen A, Jonassen I, Thompson EM, Troed- filtering for high-throughput amplicon-sequencing. Nucleic Acids sson C (2014) Characterization of the 18S rRNA gene for designing Research 43: 2513–2524. https://doi.org/10.1093/nar/gkv107 universal eukaryote specific primers. PLoS ONE 9: e87624. https:// Felsenstein J (1985) Confidence limits on phylogenies: -an ap doi.org/10.1371/journal.pone.0087624 proach using the bootstrap. Evolution 39: 783–791. https://doi. Henrichs DW, Scott PS, Steidinger KA, Errera RM, Abraham A, org/10.1111/j.1558-5646.1985.tb00420.x Campbell L (2013) Morphology and phylogeny of Prorocentrum Fonseca VG, Carvalho GR, Sung W, Johnson HF, Power DM, Neill texanum sp. nov. (Dinophyceae): a new toxic dinoflagellate from SP, Packer M, Blaxter ML, Lambshead PJD, Thomas WK, Creer S the Gulf of Mexico coastal waters exhibiting two distinct morphol- (2010) Second-generation environmental sequencing unmasks ma- ogies. Journal of Phycology 49: 143–155. https://doi.org/10.1111/ rine metazoan biodiversity. Nature Communications 1: 98. https:// jpy.12030 doi.org/10.1038/ncomms1095 Herrera-Sepúlveda A, Hernandez-Saavedra NY, Medlin LK, West N Fonseca VG, Carvalho GR, Nichols B, Quince C, Johnson HF, Neill (2013) Capillary electrophoresis finger print technique (CE-SSCP): SP, Lambshead JD, Thomas WK, Power DM, Creer S (2014) Meta- an alternative tool for the monitoring activities of HAB species in genetic analysis of patterns of distribution and diversity of marine Baja California Sur Costal. Environmental Science and Pollution Re- meiobenthic eukaryotes. Global Ecology and Biogeography 23: search 20: 6863–6871. https://doi.org/10.1007/s11356-012-1033-7 1293–1302. https://doi.org/10.1111/geb.12223 Holmes MJ (1998) Gambierdiscus yasumotoi sp. nov. (Dinophyce- Gao Y, Fang H, Dong Y, Li H, Pu C, Zhan A (2017) An improved meth- ae), a toxic benthic dinoflagellate from southeastern Asia. - Jour od for the molecular identification of single dinoflagellate cysts. nal of Phycology 34: 661–668. https://doi.org/10.1046/j.1529- PeerJ 5: e3224. https://doi.org/10.7717/peerj.3224 8817.1998.340661.x Guillerault N, Bouletreau S, Iribar A, Valentini A, Santoul F (2017) Holmes MJ, Teo SLM (2002) Toxic marine dinoflagellates in Singa- Application of DNA metabarcoding on faeces to identify European pore waters that cause seafood poisonings. Clinical and Experi- catfish Silurus glanis diet. Journal of Fish Biology 90: 2214–2219. mental Pharmacology and Physiology 29: 829–836. https://doi. https://doi.org/10.1111/jfb.13294 org/10.1046/j.1440-1681.2002.03724.x

https://mbmg.pensoft.net 12 Sze et al.: Characterising planktonic dinoflagellate diversity in Singapore using DNA metabarcoding

Huelsenbeck JP, Ronquist F (2001) MRBAYES: Bayesian inference ferred from nuclear-encoded small subunit ribosomal DNA. Protist of phylogenetic trees. Bioinformatics 17: 754–755. https://doi. 156: 393–398. https://doi.org/10.1016/j.protis.2005.09.002 org/10.1093/bioinformatics/17.8.754 Kuwahara VS, Leong SCY (2015) Spectral fluorometric characteriza- John U, Medlin LK, Groben R (2005) Development of specific rRNA tion of phytoplankton types in the tropical coastal waters of Sin- probes to distinguish between geographic clades of the Alexandri- gapore. Journal of Experimental Marine Biology and Ecology 466: um tamarense species complex. Journal of Plankton Research 27: 1–8. https://doi.org/10.1016/j.jembe.2015.01.015 199–204. https://doi.org/10.1093/plankt/fbh160 Landsberg JH, Flewelling LJ, Naar J (2009) Karenia brevis red tides, John U, Litaker RW, Montresor M, Murray S, Brosnahan ML, Anderson brevetoxins in the food web, and impacts on natural resources: DM (2014) Formal revision of the Alexandrium tamarense species decadal advancements. Harmful Algae 8: 598–607. https://doi. complex (Dinophyceae) taxonomy: the introduction of five species org/10.1016/j.hal.2008.11.010 with emphasis on molecular-based (rDNA) classification. Protist Le Bescot N, Mahé F, Audic S, Dimier C, Garet M-J, Poulain J, Winck- 165: 779–804. https://doi.org/10.1016/j.protis.2014.10.001 er P, de Vargas C, Siano R (2016) Global patterns of pelagic di- John U, Tillmann U, Hülskötter J, Alpermann TJ, Wohlrab S, Van de noflagellate diversity across protist size classes unveiled- byme Waal DB (2015) Intraspecific facilitation by allelochemical mediat- tabarcoding. Environmental Microbiology 18: 609–626. https://doi. ed grazing protection within a toxigenic dinoflagellate population. org/10.1111/1462-2920.13039 Proceedings of the Royal Society B—Biological Sciences 282: Leblanc K, Quéguiner B, Diaz F, Cornet V, Michel-Rodriguez M, de 20141268. https://doi.org/10.1098/rspb.2014.1268 Madron XD, Bowler C, Malviya S, Thyssen M, Grégori G, Rem- Jung SW, Joo HM, Park JS, Lee JH (2010) Development of a rapid and bauville M, Grosso O, Poulain J, de Vargas C, Pujo-Pay M, Conan P effective method for preparing delicate dinoflagellates for scanning (2018) Nanoplanktonic diatoms are globally overlooked but play a electron microscopy. Journal of Applied Phycology 22: 313–317. role in spring blooms and carbon export. Nature Communications 9: Karlson B, Godhe A, Cusack C, Bresnan E (2010) Introduction to meth- 953. https://doi.org/10.1038/s41467-018-03376-9 ods for quantitative phytoplankton analysis. In: Karlson B, Cusack Leblond JD, Lasiter AD, Li C, Logares R, Rengefors K, Evens TJ C, Bresnan E (eds) Microscopic and molecular methods for quan- (2010) A data mining approach to dinoflagellate clustering accord- titative phytoplankton analysis. UNESCO, Paris, 5–12. https://doi. ing to sterol composition: correlations with evolutionary history. In- org/10.1007/s10811-009-9461-6 ternational Journal of Data Mining and Bioinformatics 4: 431–451. Katoh K, Standley DM (2013) MAFFT multiple sequence alignment https://doi.org/10.1504/IJDMB.2010.034198 software version 7: improvements in performance and usability. Mo- Lee AC, Baula IU, Miranda LN, Sin TM (2015) A photographic guide lecular Biology and Evolution 30: 772–780. https://doi.org/10.1093/ to the marine algae of Singapore. Tropical Marine Science Institute, molbev/mst010 National University of Singapore, Singapore, 1–201. Katoh K, Toh H (2008) Recent developments in the MAFFT multiple Leong SCY, Lim LP, Chew SM, Kok JWK, Teo SLM (2015) Three new sequence alignment program. Briefings in Bioinformatics 9: 286–298. records of dinoflagellates in Singapore’s coastal waters, with- ob https://doi.org/10.1093/bib/bbn013 servations on environmental conditions associated with microalgal Katoh K, Misawa K, Kuma K, Miyata T (2002) MAFFT: a novel growth in the Johor Straits. Raffles Bulletin of Zoology S31: 24–36. method for rapid multiple sequence alignment based on fast Fou- Lim HC, Leaw CP, Tan TH, Kon NF, Yek LH, Hii KS, Teng ST, Razali rier transform. Nucleic Acids Research 30: 3059–3066. https://doi. RM, Usup G, Iwataki M, Lim PT (2014) A bloom of Karlodinium org/10.1093/nar/gkf436 australe (Gymnodiniales, Dinophyceae) associated with mass mor- Kazemi Z, Hashim NB, Aslani H, Liu Z, Craig PM, Chung DH, Ismail tality of cage-cultured fishes in West Johor Strait, Malaysia. Harmful M (2014) Calibration of hydrodynamic modeling in western part of Algae 40: 51–62. https://doi.org/10.1016/j.hal.2014.10.005 Johor Strait, Malaysia. International Journal of Environmental Re- Lin S, Zhang H, Hou Y, Miranda L, Bhattacharya D (2006) Develop- search 8: 591–600. ment of a dinoflagellate-oriented PCR primer set leads to- detec Ki JS (2012) Hypervariable regions (V1–V9) of the dinoflagellate 18S tion of picoplanktonic dinoflagellates from Long Island Sound. rRNA using a large dataset for marker considerations. Journal of Applied Applied Environmental Microbiology 72: 5626–5630. https://doi. Phycology 24: 1035–1043. https://doi.org/10.1007/s10811-011-9730-z org/10.1128/AEM.00586-06 Kok SP, Kikuchi T, Toda T, Kurosawa N (2012) Diversity analysis of Lin S, Zhang H, Hou Y, Zhuang Y, Miranda L (2009) High-level diversi- protistan microplankton in Sagami Bay by 18S rRNA gene clone ty of dinoflagellates in the natural environment, revealed by assess- analysis using newly designed PCR primers. Journal of Oceanogra- ment of mitochondrial cox1 and cob genes for dinoflagellate DNA phy, 68: 599–613. https://doi.org/10.1007/s10872-012-0121-0 barcoding. Applied Environmental Microbiology 75: 1279–1290. Kok SP, Kikuchi T, Toda T, Kurosawa N (2012) Diversity and commu- https://doi.org/10.1128/AEM.01578-08 nity dynamics of protistan microplankton in Sagami Bay revealed Littman RA, van Oppen MJH, Willis BL (2008) Methods for sampling by 18S rRNA gene clone analysis. Plankton and Benthos Research free-living Symbiodinium (zooxanthellae) and their distribution and 7: 75–86.https://doi.org/10.3800/pbr.7.75 abundance at Lizard Island (Great Barrier Reef). Journal of Ex- Kok SP, Kikuchi T, Toda T, Kurosawa N (2014) The protistan micro- perimental Marine Biology and Ecology 364: 48–53. https://doi. plankton community along the Kuroshio Current revealed by 18S org/10.1016/j.jembe.2008.06.034 rRNA gene clone analysis: a case study of the differences in distri- Logares R, Audic S, Bass D, Bittner L, Boutte C, Christen R, Claverie bution interplay with ecological variability. Plankton and Benthos JM, Decelle J, Dolan JR, Dunthorn M, Edvardsen B, Gobet A, Koo- Research 9: 71–82. https://doi.org/10.3800/pbr.9.71 istra WH, Mahé F, Not F, Ogata H, Pawlowski J, Pernice MC, Ro- Kühn SF, Medlin LK (2005) The systematic position of the parasitoid mac S, Shalchian-Tabrizi K, Simon N, Stoeck T, Santini S, Siano R, marine dinoflagellate Paulsenella vonstoschii (Dinophyceae) in- Wincker P, Zingone A, Richards TA, de Vargas C, Massana R (2014)

https://mbmg.pensoft.net Metabarcoding and Metagenomics 2: e25136 13

Patterns of rare and abundant marine microbial eukaryotes. Current Ronquist F, Teslenko M, van der Mark P, Ayres DL, Darling A, Höhna S, Biology 24: 813–821. https://doi.org/10.1016/j.cub.2014.02.050 Larget B, Liu L, Suchard MA, Huelsenbeck JP (2012) MrBayes 3.2: Logares R, Mangot JF, Massana R (2015) Rarity in aquatic microbes: Efficient Bayesian phylogenetic inference and model choice across placing protists on the map. Research in Microbiology 166: 831–841. a large model space. Systematic Biology 61: 539–542. https://doi. https://doi.org/10.1016/j.resmic.2015.09.009 org/10.1093/sysbio/sys029 Lopez-García P, Rodríguez-Valera F, Pedros-Alio C, Moreira D (2001) Sauvadet AL, Gobet A, Guillou L (2010) Comparative analysis between Unexpected diversity of small eukaryotes in deep-sea Antarctic protist communities from the deep‐sea pelagic ecosystem and spe- plankton. Nature 409: 603–607. https://doi.org/10.1038/35054537 cific deep hydrothermal habitats. Environmental Microbiology 12: McNamee SE, Medlin LK, Kegel J, McCoy GR, Raine R, Barra L, 2946–2964. https://doi.org/10.1111/j.1462-2920.2010.02272.x Ruggiero MV, Kooistra WHCF, Montresor M, Hagstrom J, Blan- Schnell IB, Bohmann K, Gilbert MTP (2015) Tag jumps illuminated co EP, Graneli E, Rodríguez F, Escalera L, Reguera B, Dittami S, – reducing sequence-to-sample misidentifications in metabarcoding Edvardsen B, Taylor J, Lewis JM, Pazos Y, Elliott CT, Campbell studies. Molecular Ecology Resources 15: 1289–1303. https://doi. K (2016) Distribution, occurrence and biotoxin composition of the org/10.1111/1755-0998.12402 main shellfish toxin producing microalgae within European waters: Searle D, Sible E, Cooper A, Putonti C (2016) 18S rDNA dataset profiling A comparison of methods of analysis. Harmful Algae 55: 112–120. microeukaryotic populations within Chicago area nearshore waters. https://doi.org/10.1016/j.hal.2016.02.008 Data in Brief 6: 526–529. https://doi.org/10.1016/j.dib.2015.12.042 Medlin L (2013) Molecular tools for monitoring harmful algal blooms. Shokralla S, Spall JL, Gibson JF, Hajibabaei M (2012) Next-genera- Environmental Science and Pollution Research 20: 6683–6685. tion sequencing technologies for environmental DNA research. https://doi.org/10.1007/s11356-012-1195-3 Molecular Ecology 21: 1794–1805. https://doi.org/10.1111/j.1365- Metzker ML (2010) Sequencing technologies — the next generation. 294X.2012.05538.x Nature Reviews Genetics 11: 31–46. https://doi.org/10.1038/nrg2626 Sin TM, Ang HP, Buurman J, Lee AC, Leong YL, Ooi SK, Stein- Moon-van der Staay SY, Watchter RD, Vaulot D (2001) Oceanic 18S berg P, Teo SLM (2016) The urban marine environment of Singa- rDNA sequences from picoplankton reveal unsuspected eukaryotic pore. Regional Studies in Marine Science 8: 331–339. https://doi. diversity. Nature 409: 607–610. https://doi.org/10.1038/35054541 org/10.1016/j.rsma.2016.01.011 Mordret S, Piredda R, Vaulot D, Montresor M, Kooistra WHCF, Sarno Sinigalliano CD, Winshell J, Guerrero MA, Scorzetti G, Fell JW, Ea- D (2018) DinoREF: a curated dinoflagellate (Dinophyceae) reference ton RW, Brand L, Rein KS (2009) Viable cell sorting of dinoflagel- database for the 18S rRNA gene. Molecular Ecology Resources. lates by multiparametric flow cytometry. Phycologia 48: 249–257. https://doi.org/10.1111/1755-0998.12781 https://doi.org/10.2216/08-51.1 Nagai S, Yamamoto K, Hata N, Itakura S (2012) Study of DNA ex- Spector DL (1984) Dinoflagellates: an introduction. In: Spector DL traction methods for use in loop-mediated isothermal amplification (Ed.) Dinoflagellates. Academic Press, San Diego, 1–15.https://doi. detection of single resting cysts in the toxic dinoflagellates Alex- org/10.1016/B978-0-12-656520-1.50005-5 andrium tamarense and A. catenella. Marine Genomics 7: 51–56. Srivathsan A, Sha JCM, Vogler AP, Meier R (2015) Comparing the https://doi.org/10.1016/j.margen.2012.03.002 effectiveness of metagenomics and metabarcoding for diet anal- Netzel H, Dürr G (1984) Dinoflagellate cell cortex. In: Spector DL (Ed.) ysis of a leaf-feeding monkey (Pygathrix nemaeus). Molecular Dinoflagellates. Academic Press, San Diego, 43–105. https://doi. Ecology Resources 15: 250–261. https://doi.org/10.1111/1755- org/10.1016/B978-0-12-656520-1.50007-9 0998.12302 Nguyen NH, Smith D, Peay K, Kennedy P (2015) Parsing ecological Stamatakis A (2006) RAxML-VI-HPC: Maximum likelihood-based signal from noise in next generation amplicon sequencing. New phylogenetic analyses with thousands of taxa and mixed models. Phytologist 205: 1389–1393. https://doi.org/10.1111/nph.12923 Bioinformatics 22: 2688–2690. https://doi.org/10.1093/bioinfor- Noor NM, Anton A, Amin NM (2007) Biodiversity of dinoflagellates in matics/btl446 the coastal waters off Malacca, Peninsular Malaysia. Borneo Science Stamatakis A (2014) RAxML version 8: a tool for phylogenetic analysis 21: 12–18. and post-analysis of large phylogenies. Bioinformatics 30: 1312–1313. Oldach DW, Delwiche CF, Jakobsen KS, Tengs T, Brown EG, Kemp- https://doi.org/10.1093/bioinformatics/btu033 ton JW, Schaefer EF, Bowers HA, Glasgow HB Jr, Burkholder Stamatakis A, Hoover P, Rougemont J (2008) A rapid bootstrap algo- JM, Steidinger KA, Rublee PA (2000) Heteroduplex mobility as- rithm for the RAxML web servers. Systematic Biology 57: 758–771. say-guided sequence discovery: elucidation of the small subunit https://doi.org/10.1080/10635150802429642 (18S) rDNA sequences of Pfiesteria piscicida and related dinofla- Starr M, Lair S, Michaud S, Scarratt M, Quilliam M, Lefaivre D, Robert gellates from complex algal culture and environmental sample DNA M, Wotherspoon A, Michaud R, Ménard N, Sauvé G, Lessard S, pools. Proceedings of the National Academy of Sciences USA 97: Béland P, Measures L (2017) Multispecies mass mortality of ma- 4303–4308. https://doi.org/10.1073/pnas.97.8.4303 rine fauna linked to a toxic dinoflagellate bloom. PLoS ONE 12: Posada D (2008) jModelTest: Phylogenetic model averaging. Molecu- e0176299. https://doi.org/10.1371/journal.pone.0176299 lar Biology and Evolution 25: 1253–1256. https://doi.org/10.1093/ Stern RF, Horak A, Andrew RL, Coffroth MA, Andersen RA, Küpper molbev/msn083 FC, Jameson I, Hoppenrath M, Véron B, Kasai F, Brand J, James Rambaut A, Suchard MA, Xie D, Drummond AJ (2014) Tracer: MCMC ER, Keeling PJ (2010) Environmental barcoding reveals massive Trace Analysis Tool. Version 1.6. Available from: http://beast.bio. dinoflagellate diversity in marine environments. PLoS ONE 5: ed.ac.uk/Tracer e13991. https://doi.org/10.1371/journal.pone.0013991 Ronquist F, Huelsenbeck JP (2003) MrBayes 3: Bayesian phylogenet- Stern RF, Andersen RA, Jameson I, Küpper FC, Coffroth MA, Vaulot D, ic inference under mixed models. Bioinformatics 19: 1572–1574. Le Gall F, Véron B, Brand JJ, Skelton H, Kasai F, Lilly EL, Keeling https://doi.org/10.1093/bioinformatics/btg180 PJ (2012) Evaluating the ribosomal internal transcribed spacer (ITS)

https://mbmg.pensoft.net 14 Sze et al.: Characterising planktonic dinoflagellate diversity in Singapore using DNA metabarcoding

as a candidate dinoflagellate barcode marker. PLoS ONE 7: e42780. 18S rRNA gene with a focus on photosynthetic groups and especial- https://doi.org/10.1371/journal.pone.0042780 ly Chlorophyta. Environmental Microbiology 20: 506–520. https:// Strassert JFH, Karnkowska A, Hehenberger E, del Campo J, Kolisko M, doi.org/10.1111/1462-2920.13952 Okamoto N, Burki F, Janouškovec J, Poirier C, Leonard G, Hallam Treonis AM, Unangst SK, Kepler RM, Buyer JS, Cavigelli MA, Mirsky SJ, Richards TA, Worden AZ, Santoro AE, Keeling PJ (2018) Sin- SB, Maul JE (2018) Characterization of soil nematode communi- gle cell genomics of uncultured marine alveolates shows paraphy- ties in three cropping systems through morphological and DNA ly of basal dinoflagellates. ISME Journal 12: 304–308. https://doi. metabarcoding approaches. Scientific Reports 8: 2004. https://doi. org/10.1038/ismej.2017.167 org/10.1038/s41598-018-20366-5 Swofford DL (2003) PAUP*. Phylogenetic Analysis Using Parsimony Usup G, Leaw CP, Ahmad A, Lim PT (2002) Phylogenetic relationship (*and Other Methods). Version 4. http://paup.csit.fsu.edu of Alexandrium tamiyavanichii (Dinophyceae) to other Alexandrium Taberlet P, Coissac E, Pompanon F, Brochmann C, Willerslev E (2012) species based on ribosomal RNA gene sequences. Harmful Algae 1: Towards next-generation biodiversity assessment using DNA 59–68. https://doi.org/10.1016/S1568-9883(02)00003-3 metabarcoding. Molecular Ecology 21: 2045–2050. https://doi. Wang H, Lu D, Huang H, Göbel J, Dai X, Xia P (2011) First observation org/10.1111/j.1365-294X.2012.05470.x of Karlodinium veneficum from the East China Sea and the coastal Takabayashi M, Adams LM, Pochon X, Gates RD (2012) Genetic di- waters of Germany. Acta Oceanologica Sinica 30: 112–121. https:// versity of free-living Symbiodinium in surface water and sediment doi.org/10.1007/s13131-011-0168-6 of Hawai‘i and Florida. Coral Reefs 31: 157–167. https://doi. Yamaguchi A, Horiguchi T (2005) Molecular phylogenetic study of the org/10.1007/s00338-011-0832-5 heterotrophic dinoflagellate genus Protoperidinium (Dinophyceae) Takano Y, Horiguchi T (2004) Surface ultrastructure and molecu- inferred from small subunit rRNA gene sequences. Phycological lar phylogenetics of four unarmored heterotrophic dinoflagel- Research 53: 30–42. https://doi.org/10.1111/j.1440-1835.2005. lates, including the type species of the genus Gyrodinium (Di- tb00355.x nophyceae). Phycological Research 52: 107–116. https://doi. Yamamoto S, Masuda R, Sato Y, Sado T, Araki H, Kondoh M, Minamo- org/10.1111/j.1440-1835.2004.tb00319.x to T, Miya M (2018) Environmental DNA metabarcoding reveals lo- Tan CT, Lee EJ (1986) Paralytic shellfish poisoning in Singapore. An- cal fish communities in a species-rich coastal sea. Scientific Reports nals of the Academy of Medicine, Singapore 15: 77–79. 7: 40368. https://doi.org/10.1038/srep40368 Tang YZ, Kong L, Holmes MJ (2007) Dinoflagellate Alexandrium Zardoya R, Costas E, López-Rodas V, Garrido-Pertierra A, Bautista JM leei (Dinophyceae) from Singapore coastal waters produces a wa- (1995) Revised dinoflagellate phylogeny inferred from molecular ter-soluble ichthyotoxin. Marine Biology 150: 541–549. https://doi. analysis of large-subunit ribosomal RNA gene sequences. Jour- org/10.1007/s00227-006-0396-z nal of Molecular Evolution 41: 637–645. https://doi.org/10.1007/ Tanzil JTI, Ng PKA, Tey YQ, Tan HYB, Yun YE, Huang D (2016) A BF00175822 preliminary characterisation of Symbiodinium diversity in some Zhang J, Kobert K, Flouri T, Stamatakis A (2014) PEAR: a fast and common corals from Singapore. COSMOS 12: 15–27. https://doi. accurate Illumina Paired-End reAd mergeR. Bioinformatics 30: org/10.1142/S0219607716500014 614–620. https://doi.org/10.1093/bioinformatics/btt593 Taylor FJR, Hoppenrath M, Saldarriaga JF (2008) Dinoflagellate diversity and distribution. Biodiversity Conservation 17: 407–418. https://doi.org/10.1007/s10531-007-9258-3 Supplementary material 1 Tham AK (1953) A preliminary study of the physical, chemical and biological characteristics of Singapore Straits. HM Stationery Off 8-bp barcode and primer sequences used for multiplexed sequencing (London): 1–65. Tham AK (1973) Seasonal distribution of the plankton in Singapore Authors: Yue Sze, Lilibeth N. Miranda, Tsai Min Sin, Danwei straits. Special Publication dedicated to NK Panikkar, Marine Bio- Huang logical Association of India 60: 60–73. Data type: Sequencing data Copyright notice: This dataset is made available under the Thomsen PF, Kielgast J, Iversen LL, Møller PR, Rasmussen M, Will- Open Database License (http://opendatacommons.org/licens- erslev E (2012) Detection of a diverse marine fish fauna using en- es/odbl/1.0/). The Open Database License (ODbL) is a license vironmental DNA from seawater samples. PLoS ONE 7: e41732. agreement intended to allow users to freely share, modify, and https://doi.org/10.1371/journal.pone.0041732 use this Dataset while maintaining this same freedom for oth- Tragin M, Zingone A, Vaulot D (2018) Comparison of coastal phyto- ers, provided that the original source and author(s) are credited. plankton composition estimated from the V4 and V9 regions of the Link: https://doi.org/10.3897/mbmg.2.245136.suppl1

https://mbmg.pensoft.net